'The end of statistics?'
09 February 2017

A very thought provoking article with this title was published in the Guardian in the Journal section on 19 January by William Davies. This section is sub-titled the long read, giving authors chance to present a very detailed perspective on their theme. 

Davies provides a very detailed analysis of the current crisis facing statistics, and the growing lack of trust in them as a source of truth, especially in the political sphere. Statistics are suffering a decline in authority, as are the experts who analyse them in the ‘post-truth’ world. 

As Davies states: ‘Either the state continues to make claims that it believes to be valid and is accused by sceptics of propaganda, or else, politicians and officials are confined to saying what feels plausible and intuitively true, but may ultimate be inaccurate. Either way, politics becomes mired in accusations of lies and cover ups’.

Facts v feelings

As Davies describes, the antipathy runs deep, especially with those of a far-right persuasion, and we market researchers are part of the supposed army of economists and statisticians rejected by voters as being untrustworthy and somehow arrogant in their attempts to inform public debate. 

The public also feel that statistics, numbers, do not represent them and their experience of the world. For example, whilst data on the number of people in employment is collected, and now also on underemployment, politicians tend to point to the former, whilst for many people in today’s ‘gig economy’ the latter is the crucial issue. Similarly, polling provides information on what the public feel about issues of the day, but it is the application of statistics that provides the output. 

However, simply collecting data on likelihood to vote provides nothing on the emotional intensity people really feel about turning out to vote on the day. However, as Davies argues many people respond warmly to qualitative evidence – they may reject the data on immigration levels as untrustworthy, but are more empathetic to stories of the plight faced by an individual migrant and their family. Politics grounded on statistics is ‘elitist, undemocratic and oblivious to people’s emotional investments in their community and nation’ - the politics of facts versus one based on emotions. 

Politicians and technocrats have lost touch with what it feels like to be an individual citizen. It’s that quant v qual issue, facts versus feelings, as will be debated at a session at the MRS Annual Conference - Impact 2017 next month (see my last blog on our latest Landmark Paper for more on this). 

However, as Davies argues, the history of statistics since its emergence in the second half of the 17^th century as a key facilitator of a new government perspective focussing on demographic trends demonstrates how they have enabled people to understand society – the research undertaken by The Joseph Rowntree Foundation and the national Census, being two long-lived examples in the UK. ‘Statistics would do for population what cartography did for territory’.

Behind the curve?

Davies argues that the inability of statisticians to keep pace with the changes in society over recent decades lies at the heart of the current crisis. National figures fail to identify the polarisations that exist within societies – successful economic centres, such as London, are a world away from the plight of ‘rust-belt’ communities in other parts of the UK when politicians take a macro, upbeat view of the overall British economy, using indicators such as GDP and employment levels. We discussed this in the context of social class at one of last year’s IJMR Lectures.

Privatisation of statistics

A further threat to trust is the shift from statistics emanating from public sources to ones collected by commercial organisations. Here, we have often no knowledge they are being collected in the first place, or what they say about us, ‘big data’ representing a different type of knowledge, with a new mode of expertise – data is captured first and the research questions come later. Confidentiality is the key to creating commercial advantage. There is no equivalent of the Office of National Statistics to put a seal of approval on them. Facebook, for example, can analyse the data on billions of people, but has no incentive to reveal the findings. 

There is also the rise of a whole sector devoted to profiling millions of Americans (and in other countries?) to create targets for political campaigning. Again, the findings remain secret. All this sits uncomfortably with campaigns, such as the Open Data Institute in the UK and MyData, to make data publicly available or give citizens a right of ownership over the data they generate. So, one of the most worrying aspects of a post-statistical society is the privatisation of statistics.

Regaining trust

Davis concludes that the battleground ‘is not between an elite-led politics of facts versus a populist politics of feeling. It is between those still committed to public knowledge and public argument and those who profit from the ongoing disintegration of these things’. As market researchers, we are at the centre of the storm, viewed as using our (derided?) expertise to provide the statistics used by politicians that are perceived by citizens to be painting a very different picture of the world as they see it, with little compassion for their plight. We also know that we have encountered an increasing challenge in predicting voter behaviour. We are also part of the new ‘big data’ world where statistics are commercialised by large international corporates and hidden from view, leading to anger and distrust with decisions based on automated systems or where privacy concerns go unaddressed. 

However, I’m sure that at Impact 2017 we will see many very positive stories where our work has led to fresh insights that have led to societal challenges being addressed in innovative ways, where real people having benefitted from our expertise in designing research that really gets to the heart of a problem – whether quant or qual, or preferably a blend of both to use the latter to add depth and emotion to the former. We also need to ensure that we do keep pace with the rapid changes taking place in our societies and do all we can to guide clients, whether in the public or private sectors, to recognise the need to view things differently. 

We are proud of our evidence-based traditions, but we have to win over an increasingly sceptical public by finding creative ways to regain and retain their trust.